Enhanced solutions for misuse network intrusion detection system using SGA and SSGA

العناوين الأخرى

الحلول المحسنة لنظام كشف التطفل باستخدام الخوارزمية الجينية البسيطة و الخوارزمية الجينية الحالة المستقرة

مقدم أطروحة جامعية

Jabr, Sabah Abd al-Aziz

مشرف أطروحة جامعية

Nasir al-Din, Hibah H. O.

أعضاء اللجنة

Uways, Suhayl
al-Husayni, Muhammad

الجامعة

جامعة الشرق الأوسط

الكلية

كلية تكنولوجيا المعلومات

القسم الأكاديمي

قسم نظم المعلومات الحاسوبية

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2015

الملخص الإنجليزي

One of the most widely acknowledged purposes of using the Internet is data transfer; it is an essential way of communicating personal and sensitive data.

Therefore, the need for protecting such data against hackers and intruders is at most importance.

Many security systems were built for this purpose; Intrusion Detection Systems (IDS) are one of those systems.

The main function of Intrusion Detection System is to monitor the incoming connections and detect attacks.

The purpose of this thesis is to verify the power of Simple Genetic Algorithm (SGA) versus Steady State Genetic Algorithm (SSGA) in intrusion detection field.

This is achieved by developing two models of IDS.

In the first model, the Simple Genetic Algorithm was used to build IDS (SGA based IDS), while in the second model; Steady State Genetic Algorithm was used to build IDS (SSGA based IDS).

The evaluations and the experiments were performed using the NSL-KDD intrusion detection dataset.

The experimental results demonstrated that performing an IDS using SGA gives higher performance results than using SSGA according to the value of Detection rate (DR) where it achieved an average of DR equal to 88.5%, while SSGA based IDS achieved an average of DR equal to 72.53%.

Also the number of the new generated rules using SGA is more than the number of the new generated rules using SSGA, despite that, the training time for SGA experiments is shorter than the training time for SSGA.

On other hand, SSGA based IDS produced an average of False Positive Rate (FPR) equal to 4.66% which is considered relatively better than SGA based IDS that produced an average of FPR equal to 5.21%.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

عدد الصفحات

75

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-694619